[Zhangshi]

Dear Robin，

After the discussion with you the day before yesterday，I found some problems.I can‘t get any data I wanted.I thinks maybe I misunderstand the meanning of charU. When I set datas for the program.I always calculate the max velocity of the particle by stokes drag force formula at first.Then I set the charU to the value which is aqual to the max velocity of particle.I think taht’s not right.How should I set for the charU?

Best
Zhangshi

[Zhangshi]

Dear Robin，

Thanks for your share.Maybe I think the less value of episilon is，the more similar to sphere the particle will be.I’ll try to understand this point.

Best
Zhangshi

[Zhangshi]

Dear Robin，

I forget to tell you that I set episilon for about 1E-7m，I think it’s more similar to the sphere.Or maybe it’s not really important.
I’m waiting for your answer.Thank you so much.

Best
Zhangshi

[Zhangshi]

Dear Robin，

I got it!My settling of the particle size are 40μm，60μm and 80μm.Maybe the lattice is a little large.Thanks for your suggestions.
By the way,I wonder if the value of the relaxation time is necessary.Does it have a certain range?
And can I set no-slip boundary wall with the situation that I create a
larger area？

Best
Zhangshi

[Zhangshi]

Dear robin，

Thanks for your prompt answer.
I create a cube area in size of 600μm*600μm*720μm，and the size in lattice is50*50*60.The density of fluid and particle is 7900 and 3900 in kg/m^3.
Here are some parameters.
LatticeU=0.000904
tau=0.797
charL=0.0006 m
charU=0.00054 m/s

Conversion factors：
phyDeltaX=1.2e-05 m
phyDeltaT=2e-05 s
phyVelocity=0.5973 m/s
I set no-slip boundary condition for the wall.

Best
Zhangshi

[Zhangshi]

Dear Robin，

Thanks for all your help.I’m sorry to reply so late.Yesterday I’ve tried
to modify the source code.But maybe it’s too hard for me now.Maybe I’ll try another way.
Thank you again.

Best
Zhangshi

[Zhangshi]

Dear robin，

Thanks for your patient answer.I’m sorry for asking for advice too often.
It seems a hard work for me to make an attempt on creating a new SmoothIndicator,but I’ll think about trying this way.You suggest that I shoulduse an existing SmoothIndicator as a blueprint.
Do you mean that I should just use some simple smoothIndicators for simulation,such as smoothIndicatorCuboid? Or try to combine these geometric primitives with a more complex one?
By the way,I’ve build a oil-water surface with the help of you.And I was baffled by how to set the change of the acceleration when the particle arrived the oil.
You are so kind for answering so many questions for me.I don’t know how to express my gratitude.If I could post my problems more easier on forum,I’ll have more time to solve problems by myself with your guidance,rather than just ask for advice.

Best
Zhangshi

[Zhangshi]

Dear Nicolas,

Thanks for your detailed answer.Though it will take a little more for me to understand 1b.I’ll read the doxygen again.
By the way,have you ever tried the STLreader to add a particle based on your own stl files?Is this feasible?
Thank you again for your help.

Best
Zhangshi

[Zhangshi]

Dear Robin，
Thanks for your reply.For first problem，it seems that the RPM in my computer is not really enough.I‘ll change another one.
For stl files,I have some trouble for code programming.
I read the user guide.But I only found the usage of stlreading in geometry preparing.I have no idea how to use stlreading to build a particle.This is a really important and confusing point for me.Could you give me some guidance?
I wonder if the following codes are necessary for this.

IndicatorCuboid3D<T> cuboid(extend, origin);
 CuboidGeometry3D<T> cuboidGeometry(cuboid,      converter.getConversionFactorLength(), 7);
 HeuristicLoadBalancer<T> loadBalancer(cuboidGeometry);

Thanks for your patience.

Best
Zhangshi

[Zhangshi]

Dear Mathias,

I’m sorry for asking too many questions.I’m glad to going to spring school for further learning.But it seems that I don’t have enough time to get there.
Some of the questions above have been solved by myself.But there are still two problems.Do you mind listenning for a while?

First is question 2.

`AnalyticalConst3D<T,T> one ( 1. );
SmoothIndicatorSphere3D<T,T> sphere( {nx/2., nx/2., nx/2.}, radius, 10.*alpha );
AnalyticalIdentity3D<T,T> rho( one );
AnalyticalIdentity3D<T,T> phi( one – sphere – sphere );
sLattice1.iniEquilibrium( superGeometry, 1, rho, zeroVelocity );
sLattice2.iniEquilibrium( superGeometry, 1, phi, zeroVelocity );`

AnalyticalIdentity3D<T,T> phi( one - sphere - sphere );
It’s used for constructing spherical erea.But “one” is not the same kind as the sphere.And why here subtract sphere for twice?

2.Something about question 1 is really confusing.How to get the value of these variable in lattice units?Would yuo give me some advice?

Thanks for you reply!
Best
Zhangshi

• This reply was modified 3 years, 11 months ago by Zhangshi.

[Zhangshi]

Dear Robin,

Thanks for your support and guidance!You are so kind.Now I’m reading relevant codes,such as “youngLaplace3D”.I hope I can get over it soon.

Best
Zhangshi

[Zhangshi]

Dear Robin，

I’m so sorry to bother you again.I have no idea if I should choose the shan-chen model or the free energy model for simulating oil-water interface.Could you recommed some relevant code for me?
Have a nice day!

Best
Zhangshi

• This reply was modified 3 years, 11 months ago by Zhangshi.

[Zhangshi]

Dear Robin

Thanks for your inviting.I have ever thought about the SpringSchool.
But it seems that UK is too far for me to get there.I feel sorry about that.But,I’m glad to share this information with my friends.

Best reagards
Zhangshi

[Zhangshi]

Dear Robin.Trunk

Thanks for your detailed answers.Now,I’ve been more clear about my
further simulation.You really helped me a lot.But,I’m still thinking that if STLreader can perform better than the SmoothIndictorIdentity.
Whatever,it’s important for me to start the oil-water simulation.
Thank you!!

Best
Zhangshi

• This reply was modified 3 years, 11 months ago by Zhangshi.

[Zhangshi]

Dear Robin.Trunk

Thank you very much for your prompt reply.I’m sorry that I can’t reply you in time because of time difference.First,allow me to express my best respect for you.Although my English is not very good,I’ll try my best to express clearly.
As you mentioned,this time I changed the ratio of the density,and the orientation became more obvious.But the change seemed to be unreasonable.It’s necessary for me to make further adjustment on parameters.May I ask another two questions?I’m sorry to take up your time again.First I tried to build a complex floating particle by combining geometric primitives,For example,combining two or there spheres into another further complex particle.I’m not aware that if the class ”SmoothIndicatorIdentity” can help me out.And last,for further simulation,I want to simulate the particle floating through the water-oil interface,as initially the oil layer is laid on the water surface quitely.I have some ideas about building the water-oil interface,just like the example “RayleighTaylor3D”.But it’s a problem for me to consider the dynamic setting of the particle,such as acceleration.
Thank you sincerely.

[Author]

Posts